This is a master's degree project for the LBBE (Laboratoire de Biométrique et de Biologie Évolutive), made by students from the Master de bioinformatique from Lyon.

The aim of this collection of scripts is to allow the user to plot various metrics relative to the genome's composition. It allows both to build a csv database of mappings between the genes of one species and all the other in the choosen group and to download and calculate metrics on the genome of selected species.

The wrapper script is written in Snakemake. Here's the procedure to follow to install snakemake through conda and allow for optimal integration :

1. Install conda
2. Setup bioconda channels:

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge

3. Install snakemake (conda install snakemake)

If you don't intend to use snakemake, you can install manually the dependencies and use the scripts one by one. You'll need the following python packages:

• requests
• biopython
• bcbiogff

and the R package dplyr.

You'll also need:

• any annotation file for the reference species (whose genes will be searched on orthodb)
• two files from orthodb: OGs to genes correspondence and xrefs associated with Ortho DB gene from https://www.orthodb.org/?page=filelist

File paths are to be set in the config.yaml or given to the proper script.

A global interface is available through snakemake. To use the full pipeline (creation of datasets, download of genome and plotting), run snakemake -j [threads to use] --use-conda. This will create the conda envs for the pipeline and then run it with the number of cores you specified (only requests to the NCBI are multi-threaded through the threading python lib, don't hesitate to specify a high number like 10 or 20).

To use only the update part, run snakemake -j [threads to use] --use-conda update.

To use only the plotting part, just make sure that all files created by update are in the folder and run snakemake -j [threads to use] --use-conda again.

Individual scripts do the following:

• get_gene_names.py: get gene names from gff, outputs a json file containing a list of genes and a csv containing informations about genes for plotting (chromosome, start position, symbol)
• get_orthologs.py: get the orthologs of geneIDs passed as json to the script, outputs two csv, one with taxonomy information and the other with the genes/species associated with each gene from the species
• species_selection.R: select species from the file created by the preceding script in order to optimize plotting
• get_metrics_from_geneID.py: download genomes and creates metrics files.
• heatmap.R: plot values created by the preceding script as an heatmap.